EP2224301A1 - Automation assembly with an industrial automation component and method for managing an industrial automation component - Google Patents

Abstract

The automation arrangement has multiple automation components which are executed for life span and are associated with maintenance instance. The maintenance instance is aligned with a medium for detection state of the automation components, and is also aligned with another medium for influencing the automation components.

Description

The invention relates to an automation arrangement with a multiplicity of industrial automation components and to a method for managing an industrial automation component according to claim 9.

Industrial automation arrangements can be generally described as a system in which as components (automation components) sensors, actuators, operating elements, infrastructure components (routers, servers, switches and the like) are arranged in a data network (for example Profibus, fieldbus, LAN etc.).

An automation arrangement, generally also called a system, has the purpose to fulfill a specific task. For this purpose, a system can usually be dissected. The individual components, which are clearly distinguishable from each other, are the components mentioned.

A component thus fulfills certain subtasks of a system and contributes to the functionality of the system. For the system as a whole to fulfill its task, the correct interaction of all components is necessary. For the component to fulfill the task, a runtime environment or infrastructure is required.

In a run-time environment or on an infrastructure, you can have multiple identical or different components simultaneously go to execution. A built-in component becomes part of the infrastructure. To pass on the interim results of individual subtasks, the components have external interfaces.

A hierarchical structure of a system can also be achieved with this structure. For this purpose, interfaces for links can be assigned to a system itself. An active system can then be understood as a component. This nesting ensures that the user of a system remains hidden from the structure and thus the complexity of the system. By contrast, the administrator of a system can gradually ascertain the relationships and maintain the system and keep it functional.

In general, components in a system can in some cases only be used once, but often simultaneously by several other components. The more often one component is used by another, the more of the available capacity is occupied. In this case, the utilization information regarding the existing and the utilized capacity can be determined for a component. The utilization information can be determined and provided by any means (including manually).

For the structure shown here, the components also have another function at runtime: A component displays its own state and the state of the connection to the dependent components linked to it. From the outside, it is therefore possible to read at any time whether the component is working or a link is no longer functional.

It is often demanded that a system should be decentralized, ie decisions are not made by a central one Component is met, but the tasks are distributed so that can be decided locally on different components. For the present system concept, this often means that components make a decision decentrally by selecting and using a specific link. In a central approach, this would be sending the information to a central point and then being decided.

It is a problem of such decentralized structures that while the system can be managed and configured at runtime, this process is still done externally by manual intervention or by a higher-level administrative entity.

It is therefore an object of the present invention to propose an automation arrangement with industrial automation components, which should largely carry out its own configuration, monitoring and optimization.

The object is achieved by an automation arrangement according to the patent claim 1 and by a method according to the patent claim 9.

It is an essential idea of the invention that each or at least a majority of the components (automation components) is assigned its own management entity (autonomous administrator), which should be constructed largely independent of the component, so that they the state of the component and the links of the component can also record and evaluate in the event of an error. For this purpose, the management instance is equipped with sensors and actuators for access to the respective component to be managed. In a knowledge base (eg in the form of a database) is deposited, How the current state of the component is determined from the sensor values. Conversely, the knowledge base also defines how the component behaves, or how the component reacts when it is manipulated from the outside. Furthermore, the knowledge base stores which state and which behavior of the component or even of the overall system (automation arrangement) is desired when and how this goal can be achieved by manipulating the individual component. This is understood below as part of a "strategy".

1. a) Überwachung der Ausführung der Komponente (Ressourcenverbrauch, Ressourcenbedarf, Performance, Fehlermeldungen, Qualität einer Dienstausführung),
2. b) Analyse des Systemzustands der Komponente,
3. c) Stabilisierung im Problemfall,
4. d) Rekonfiguration und Wiederherstellen der Funktionalität ("Heilung" oder "Selbstheilung"),
5. e) Optimierung hinsichtlich neuer Anforderungen.

In short, the administrative entity performs the following tasks:

1. a) monitoring the execution of the component (resource consumption, resource requirements, performance, error messages, quality of service execution),
2. b) analysis of the system state of the component,
3. c) stabilization in case of problems,
4. d) reconfiguration and restoration of functionality ("healing" or "self-healing"),
5. e) Optimization for new requirements.

1. a) Beobachten (dabei werden die Sensoren einer Komponente abgefragt),
2. b) Bewerten (Ermittlung des aktuellen Zustandes der Komponente aus den ermittelten Sensorwerten und der Wissensbasis),
3. c) Bedienen (Übertragung der Ergebnisse der zweiten Phase auf die Komponente mittels der Effektoren).

The administrative entity thus performs three different steps (phases) in cyclical sequence:

1. a) monitoring (the sensors of a component are queried),
2. b) evaluating (determination of the current state of the component from the determined sensor values and the knowledge base),
3. c) operation (transfer of the results of the second phase to the component by means of the effectors).

The solution of the task provides an automation arrangement with a plurality of industrial automation components, wherein a plurality of these automation components is administrable at runtime and is each associated with an administrative entity, and wherein the management entity is equipped with first means for state detection of the automation component and with second means for influencing the automation component. In this case, the administrative entity is linked to a database as the carrier of a knowledge and strategy base, the database containing at least an indication of the task to be performed by the automation arrangement and / or an indication of a state to be envisaged for the automation arrangement, also in the sense of a strategy , and the management entity is set up for an assessment of a detected state taking into account the information from the database and a resulting influence on the respective automation component. The automation components of such an arrangement can largely autonomously, ie without user intervention, respond to changed operating conditions, tasks, goals, errors, etc. and ensure or optimize the function of the automation arrangement.

The solution of the task further provides a method for managing an automation component in an industrial automation arrangement, wherein the automation component is assigned a management instance, and wherein the management entity is equipped with first means for state detection of the automation component and with second means influencing the automation component. Furthermore, the management instance is linked to a database, wherein the database contains at least an indication of the task to be performed by the automation arrangement and / or an indication of a state to be desired for the automation arrangement or the automation component. In a first step, the Management entity a state of the automation component to be managed by means of the first means, in a second step, the management entity performs an assessment of the state information determined in the first step taking into account at least one indication from the database, and in a third step, the management entity determines from the evaluation a resulting influence on the respective automation component and applies this by means of the second means to this automation component. Due to the essentially pairwise assignment of administrative entities to automation components, the individual automation components can be optimally configured or influenced locally, whereby the task of the entire automation arrangement can be taken into account by accessing the database with local and ideally also superordinate goals or a strategy.

Advantageous embodiments of the automation arrangement according to the invention are specified in the dependent claims. The features and advantages listed here apply mutatis mutandis to the inventive method.

Advantageously, knowledge is recorded in the database as to how influencing the respective automation component changes the state of this automation component, in which case the management entity is set up to take this knowledge into account when assessing the detected state. The information recorded in the database advantageously relates to a functional group of the automation components. In this way, the behavior of the individual automation component can be aligned with the goals of a functional group or the overall arrangement.

For this purpose, the database advantageously comprises knowledge about how an influencing of the respective automation component changes an exercise of the superordinate function of the group of automation components, wherein the management entity is set up to take this knowledge into account in the evaluation. This enables the management entity to purposefully determine commands and instructions and other configuration actions and apply them to the single automation component.

A number of automation components may be based on a common infrastructure element (eg, a PC, controller, or the like); In such cases, the infrastructure element has a higher-level management instance. In this case, the higher-level administrative entity is advantageously set up to coordinate the data stored in the respective databases of the infrastructure element about the respective desired state. It advantageously has its own database with a knowledge base and can therefore use its own strategy for influencing subordinate administrative instances and thus subordinate automation components. This makes it possible to coordinate the usage of shared resources in the infrastructure element (eg, memory, power, bandwidth, computing power, etc.).

Embodiments of the inventive arrangement will be explained below with reference to the drawings. They also serve to explain the method according to the invention.

Figur 1

eine schematische Darstellung zweier miteinander vernetzter Verwaltungsinstanzen (jeweils "Autonomer Verwalter" genannt),

Figur 2

in schematischer Darstellung eine bekannte Anordnung aus 2 Personal-Computern und drei Druckern,

Figur 3

die um Verwaltungsinstanzen ergänzte Anordnung aus der Figur 2,

Figur 4

ein Fließbild einer Anordnung zum Umpumpen von Flüssigkeiten, und

Figur 5

eine schematische Darstellung der Komponenten zum Umpumpen von Flüssigkeiten.

Showing:

FIG. 1

a schematic representation of two interlinked administrative bodies (each called "Autonomer Verwalter"),

FIG. 2

a schematic representation of a known arrangement of 2 personal computers and three printers,

FIG. 3

the administrative order supplemented from the FIG. 2 .

FIG. 4

a flow diagram of an arrangement for pumping liquids, and

FIG. 5

a schematic representation of the components for pumping liquids.

For a system illustrated in the following two embodiments, an industrial automation device, hereinafter referred to as system, is additionally implemented a separate network of management entities, hereinafter referred to as autonomous manager (AV). The system itself is not changed by the Autonomous Administrators. The network of autonomous administrators has the task to carry out the self-administration, ie the configuration, monitoring and healing as well as the optimization of the respective system.

For this purpose, each industrial automation component, referred to below as the component, is assigned an autonomous administrator who exactly matches a role type of the component. An Autonomous Administrator is set up or configured depending on the component, as he must be able to capture and evaluate the state of the component and links.

The structure of two autonomous administrators is in the FIG. 1 shown.

In the FIG. 2 For example, consider a printing system as a first example. The task of the printing system is to send print jobs from PC programs to printers, which then print the print jobs. In this case, a classic printing system consists of the three role types PC program, print server and printer. Components 1 and 2 are executable PC programs (eg, a word processor) on a computer. The printers, on the other hand, are each bound to a special hardware that enables the printing of the information. The PC programs are tied to an infrastructure element in the form of a computer with an operating system and a network connection. The printers are each bound to a network-capable real printer as infrastructure elements. The infrastructure elements thus provide the components with an execution environment consisting of hardware and software resources. Added to this is the ability to communicate with the environment. A built-in component becomes part of the infrastructure.

Viewed from the inside, the printing system can be recognized as an active system with its individual components. Outwardly, however, the internal structure can no longer be seen and the printing system can be assigned as a component of other systems to the role type "printing system". A PC program can send print jobs to a color printer or a SW printer.

The concrete assignment of the components in the instance view is in the FIG. 2 to see. The print sources locally accept the print jobs on one component and pass them on to the print sinks, ie the printers. There is provided locally for printing the print job. The component print source1 is a component with the interfaces to one color and one SW printer each. In addition, the State of the component and the component made visible to the outside, so that at runtime, the state of the component can be queried. At runtime, therefore, the pressure sink A can be replaced by the pressure sinks C, since this is also of the roller type color printer. The printers also show their state to the outside, so that an assessment from the outside is possible.
On closer inspection is in the FIG. 2 to recognize that initially exist two separate printing systems, since the right part is not connected to the left by links. An advantage of the decentralized solution is thus a greater independence of the components from each other. The printing system continues to exist if the right part fails.

Print source 2 can not currently print in black and white, but color printing is done as usual. For SW printing, only a link to a SW printer, such as pressure sink B, must be established at runtime. The smallest entry point in the case of the printing system is the pressure source to a PC program.

In the FIG. 3 is shown that the pressure source is now linked to an autonomous administrator as a management entity, which ensures in a first step that appropriate printers are searched in the environment and assigned the best pressure sink depending on the strategy. This is done through the communication between the autonomous administrators. The superordinate Autonomous Administrator of the printing system merely specifies the strategy defined in the database (knowledge base) assigned to this superordinate autonomous administrator.

Examples of strategies include using the fastest printer or the printer with the largest paper supply. While the execution of the pressure source monitors the respectively assigned local Autonomous administrator this constantly. If errors occur, they will respond. For example, if a printer fails, the Autonomous Administrator will register and assign a new printer. Even without the failure of a linked component, it is regularly checked whether the assignment is favorable in the sense of the current strategy. Then, under this condition, the best component is searched for and reconnected. This can also be the original component.

The example of the printing system listed here can use a peer-to-peer system for implementation, which allows some basic services such as the search for specific components. Other approaches, however, are also possible. All three criteria of self-government such as configuration, monitoring and healing as well as optimization with the help of the autonomous administrator are fulfilled. Therefore, the present system may be referred to as an autonomous printing system.

The second example deals more closely with the circumstances of the control and automation environment. It is the control of a recirculation process, whose R & I flowchart in the FIG. 4 is shown. For the pumping process, two different types of pumps PG1 and PG2 can be used. PG1 performs the process slowly and inexpensively. PG2, on the other hand, works fast and incurs higher costs. This results in three driving styles. The first is a slow pumping, the second a fast pumping and the third is the use of both pump groups at the same time, so that a very fast pumping can be achieved.

In the FIG. 5 the structure of the control of this arrangement is shown as a system. This example shows four components with different role types. The first component is the operator control and monitoring interface (BuB interface). Three buttons are provided in it to directly select the pumping process from container B1 to container B2 in the three individual operating modes. The component thus has three interfaces to the outside.

The control component provides for each of the three driving modes corresponding control programs for execution. The two output interfaces to the role types "Transfer (slow and favorable)" and "Transfer (fast and expensive)" are used with the corresponding components PG1 ("pump group") for "pumping (slow and favorable)" and PG2 for "pumping (fast and expensive) "linked. The view in FIG. 5 thus represents the active system of the controller; Here, the decentralized at runtime manageable control can be seen.

The central component, ie the controller which controls the pumping process from B1 to B2, can easily be provided in the event of an error, since the control only has to know the pumping process as such. Which pump group is used can be specified externally at runtime. The user can dynamically choose between the three modes of operation by changing the default of the pump groups at runtime. This does not necessarily have to be the plant operator who starts over the BuB interface, but also an external algorithm or an external person such as the maintenance engineer who has taken a pump group out of service for maintenance purposes. In contrast to today's automation systems, the configuration and the design of the controller can thus be done dynamically at runtime. The smallest In the case of the control, the entry point is the command acceptance to a BuB interface.

In the FIG. 5 We can see further how the control system complements the autonomous administrators. The autonomous administrator of the BuB surface searches for a controller that can control a pumping process and links the components together. The Autonomous Umpumpen administrator now searches for a corresponding pump group.

At this point, the importance of the strategy defined in the respective knowledge base or database in the field of automation technology is clarified. The selection of the pump group depends on the global specifications of the autonomous manager of the control system. If the strategy is favorable, pump group PG1 is selected. However, if fast pumping is required PG2 is selected. If pumping is required particularly fast, both pump groups are used at the same time. But conflicts can also arise in strategies, so that compromises must be found, for example, if cheap and fast pumping is to be achieved at the same time. Solving this is the task of the Autonomous Administrators. The configurations are adapted to changed strategies during an optimization phase.

The user learns nothing of the selection problem and the strategies. He can ask for the strategy as needed or specify it himself. Ultimately, he will be able to observe a functioning Umpumpvorgang after pressing the switch. Overall, the user receives an autonomously acting system for controlling the pumping process.

Claims (10)

Automation arrangement with a large number of industrial automation components,
characterized,
that a plurality of these automation components can be managed at runtime and are each linked to an administrative instance,
that the administrative instance having first means for detecting state of the automation component and with second means is adapted for influencing the automation component,
that the management entity is linked to a database, wherein the database contains at least an indication of the task to be performed by the automation arrangement and / or an indication of a state to be sought for the automation arrangement, and
that the management entity to a valuation of a detected condition is set taking into account the indication from the database and a resulting effect on the respective automation component. Automation arrangement according to claim 1,
characterized,
that the task to be accomplished and / or to be aimed at for the automation arrangement state refers to a functional group of the automation components. Automation arrangement according to one of the preceding claims,
characterized,
that the database comprises knowledge of how influencing the respective automation component changes the state of this automation component, and
that the administrative entity is set up to take this knowledge into account when influencing the automation component. Automation arrangement according to one of the claims 2 or 3,
characterized
the database comprises knowledge of how influencing the respective automation component changes an exercise of the superordinate function of the group of automation components, and
that the administrative body is set up to take that knowledge into account in the evaluation. Automation arrangement according to one of the preceding claims,
characterized,
that a number of the automation components are based on a common infrastructure element,
wherein this infrastructure element has a higher-level management instance. Automation arrangement according to claim 5,
characterized,
in that the higher-level administrative entity is set up to coordinate the data stored in the respective databases of the infrastructure element about the respective desired state. Automation arrangement according to one of the preceding claims,
characterized,
in that the first means are sensors and / or the second means are actuators. Automation arrangement according to one of the preceding claims,
characterized,
in that the first means are interfaces for receiving and / or retrieving status data and / or the second means are interfaces for transmitting command data. Method for managing an automation component in an industrial automation system,
characterized,
that the automation component is assigned to a management entity,
that the management entity is equipped with first means for state detection of the automation component,
that the administrative instance is equipped with second means for influencing the automation component,
that the management entity is linked to a database, wherein the database contains at least an indication of the task to be performed by the automation arrangement and / or an indication of a state to be sought for the automation arrangement, and
that in a first step the management entity records a state of the automation component to be managed with the aid of the first means,
that the management entity in a second step, taking into account an indication from the database performs an evaluation of the state information,
that the administrative instance in a third step from the evaluation determines a resulting effect on the respective automation component and applies it through the second means of the automation component. Method according to claim 9,
characterized,
that the database contains at least a further indication of the automation component by the task to be accomplished and / or an indication of a aspired for the automation component state, said further indication for the evaluation is used.

Images